I am facing the same problem with prefetching data from shared mem to registers.
Did you solve this?
---
[Visit
Topic](https://discuss.tvm.apache.org/t/how-to-i-use-prefetch-with-gpu-codegen/7294/2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe
Hi,
I follow the build step in
https://discuss.tvm.apache.org/t/tvm-and-blas-libraries/4760 to use the
mkl-dnn. Then I run the script tutorials/get_started/relay_quick_start.py and
print the module ir. It seems the convolution operation not use mkldnn. After
that, I grep the src/include/pyth
Hi Experts, @Lianminzheng @jcf94
I tried to define a new operator for lstm network.The computation declaration
for lstm op has been tested and it is correct. Now I want to use
auto-scheduling to automatically generate a large search space and find a good
schedule in the space. But it can no
Hello, I've seen a few posts talking about RISC-V support, but it's not clear
to me what the current state is. Could anybody clarify to what extent RISC-V is
supported? Could you provide some pointers to repositories or documentation?
Thank you very much!
---
[Visit
Topic](https://discus
@Dileep Can you run tests/ios_rpc_test.py successfully? Your tuning script can
run against the llvm target?
Can you try the following patch and check if something changes or not?
```
diff --git a/python/tvm/autotvm/measure/local_executor.py
b/python/tvm/autotvm/measure/local_executor.py
index
Hi, everyone
For cuda target, I first fetch data from global memory to shared memory, then I
want to achieve software pipeline by prefetching some data from shared memory
to registers since shared memory request may consume tens of cycles and
sometimes even longer.
However, the underlying pr
Hi Team,
Any suggestion on the above issue .
Thanks
---
[Visit
Topic](https://discuss.tvm.apache.org/t/ios-auto-tvm-auto-tuning-was-not-happening-for-ios-metal-target/8115/2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [cl
The verification code I extracted is:
```
import logging
import numpy as np
import tvm
import random
import sys
import math
import timeit
from tvm import relay
from tvm import autotvm
def numpyBaseline(M,K,N):
np_repeat = 100
np_runing_time = timeit.timeit(setup='import numpy\n'
In my understanding, autotvm implements operators based on opreation and
schedule defination.
For example, there are pack and no pack implementations in x86/dense.py.
In the process of autotuning, the implementation-defined configs are fed into
the actual measurement, but No such correctness
this may have some help.
https://discuss.tvm.apache.org/t/how-to-retarget-tvm-to-a-new-asic-chip-as-a-device-code-generator/1276/10
https://discuss.tvm.apache.org/t/rfc-support-cce-target-name-in-tvm/922/8
https://tvm.apache.org/docs/tutorials/language/tensorize.html#sphx-glr-tutorials-langua
10 matches
Mail list logo
